Communication for Policy ImpactPAN-ALL Meeting, Penang

13 June 2009

This work was carried out with the aid of a grant from the International Development Research Centre, Canada and the Department for International Development, UK.

“Policy Impact”: what is it?

Why communicate?

Why communicate?

• Why not keep knowledge to ourselves? – We can’t bring about change alone?– Multiplier effect?

• Communication: to reach others• Communication: to make a positive impact

– A change in policy– A change in thinking

• Other reasons (or benefits of communication)?– What about to learn (from the process): e.g. web statistics

Who do/should you communicate with?

Who do you communicate with?

• Government (policy makers/regulators)• Citizens• Donors • Media• Other researchers• Private sector• Others?

E.g.1: Teleuse@BOP research change in thinking change in biz. strategy

• 6 country survey, repeated every 2 years• A few insights gained:

– “over 92% of the BOP use phones, frequently”– “the BOP use mobiles, even if they don’t own a

phone of their own”– “they are willing to spend up to USD 5 per month on

phone services, if initial barriers to entry can be lowered”

– “they are fare more likely to get information via the phone than via internet (telecenters)”

Who do we communicate findings to? In what order?

• First released to private sector– They provide the phones

• Then to policy makers & regulators– They create the enabling environment– Can provide services (e-Gov) that other’s can’t

• And to other researchers/implementers– So they design projects that are accessible by

the BOP• Then to more “academic” audiences

– To build models, enhance theoretical understanding

• Released to MediaContinuously, opportunistically

Some results

• Getting private sector telco to recognize value of BOP to company strategy– Previously focused on TOP– Later quoted saying “addressable market includes those

earning under USD 100 a month” – Based on evidence (“those @ the BOP are willing to spend up

to USD 5 per month on phone services, if initial barriers to entry can be lowered”); close to telco ARPU

• Another operator who already focused on BOP designed product– to share a mobile phone with many users and receive

payment transparently

Why include private sector? Why first? • Faster implementation/adoption [compared to govt.

response times]• Sustainable in the long term (if they can make

profits)• Better use of limited public funds

E.g. 2: Early-Warning: Change in thinking change in projects/focus

• Sri Lanka’s largest CBO, Sarvodaya– Previously focused on disaster recovery and relief

• Engagement with LIRNEasia’s “National Early Warning System: Sri Lanka” concepts

• Change in thinking: – ‘One of the biggest lessons we learned from the Tsunami

was how lacking Sri Lanka was in terms of an emergency warning system..’ (p.73), Sarvodaya, post-Tsunami report

• Change in actions– Designing projects based on disaster preparedness and

early warning

E.g.3: Mobile-Tax: Change in Policy WITHOUT change in thinking• Repealing regressive tax on mobiles SIMs in Sri Lanka

– Successful; lots of glory• But thinking within government NOT changed

– Still think mobiles as source of easy revenue (through taxation)

– Don’t account for: High tax reduced usage reduced overall revenue

– Don’t view phone as enabler of other economic activity

• Need to fight another battle the next time…

Research Impact

RESEARCHRESEARCHCOMMUNICATIONCOMMUNICATION

IMPACTIMPACT

Useful knowledge and know-how

Useful to others? Can have positive

impact on….?

What to say? To whom?How to say it? (which

formats? which media? where?)

Who says it? When?

Change Policy (one time)

Change in Thinking (long term and sustainable)

Changes in policy follows

Communication strategy and implementation

A Communication Strategy is a map

• Think of an unmapped jungle• Is it better to have an inaccurate, incomplete map

versus no map?• How often should you revise the map as you go

through the jungle?– Or should you blame the jungle for not corresponding to

the map?• Can you postpone entering the jungle until accurate

and complete maps become available?– Who will make them, if not you?

A map

• Tells you how to get from point A to point B• Good if it points to things that will help (e.g.,

sources of drinking water) or hinder (cave hic dragones)

• Best if it shows alternative paths and pros and cons

When to get a CS?• Right at the start

– Need to think of possible policy impacts and how they can be achieved, when writing the proposal

• Assume no one does basic research with IDRC money

– Only way you will obtain money for communication is if you write it into the budget

• Communication is costly• Cannot be done through volunteer labor• Rule of thumb: 10-15% of research budget needed for a

multi-country project; less if single country• IDRC is committed to funding dissemination

When do you communicate?

• When the results are in• Before the results are in, but when the time is

right

• Opportunism is an essential ingredient– Perfect is the enemy of the good– Need respond when there is demand for the story,

not when you are good and ready

Most important element of CS

• Identify your audience(s)• If more than one, assign priorities

– If all are equally important, your campaign unlikely to succeed

– Doing one thing means, having less resources for other things

• Priority can change over time– E.g., first six months senior managers of companies; after

that civil-society opinion leaders; some time down the road, academics

Choose method/media based on audience• If senior private-sector decision-makers are your audience,

succinct Powerpoint is the best bet– Certain kinds of conferences (trade events), trade journals and financial

press will increase your credibility and reinforce the message (but cannot substitute for the face-to-face)

• Face-to-face and Powerpoint best even for government– Skeptical about policy briefs being sent cold; useful in the context of a

relationship– Govt officials also impressed by your presence in certain fora– Always leave a document behind (PP or Policy Brief)

• If your audience is the general public, no alternative to TV– If opinion-leaders/subset among the public, print may work– Radio is the most difficult to work with – Having a personal brand (derived through mass media) helps

More phones in BOP households in emerging Asia than radios and computers; catching up with TVs

Source: ~10,000 sample, six-country Teleuse@BOP3 survey, October 2008

What about web presence

• Search is the new paradigm: if you can’t find an organization through web search, does it exist?

• To provide information on a demand-pull basis to institutional actors, a website is very useful

• Can be a way of advertising yourself to information brokers—important bloggers, journalists, opinion leaders

• Very harmful if you do not have it• Not necessarily beneficial by itself• Web work takes time; time that has to be taken from

something else

Don’t get hung up on hits• Look for readers who stay

– LIRNEasia average is 2:14 mts; with 70% bounce rate, that means that a significant number among the 30% stay for 4-5 mts

• Are they from among your desired audiences– Raw numbers are meaningless for strategy– Google Analytics is indispensable

Making sense of web analytics

Subcontinental Region Visits Pages/visit

Ave. time on site

% New visits

Bounce rate

S Asia 3,267 2.44 00:03:01 65.07% 67.03%

N America 1,084 1.84 00:01:50 73.43% 74.26%

N Europe 464 1.72 00:01:23 77.59% 79.31%

SE Asia 363 1.74 00:01:26 80.17% 77.13%

W Europe 275 1.55 00:00:44 84.00% 77.09%

E Asia 211 2.13 00:02:10 66.35% 67.30%

Aus NZ 201 1.68 00:00:58 86.57% 77.61%

W Asia 197 1.71 00:01:10 89.85% 78.68%

Understanding what electronic media can and cannot do

• Not direct policy influence• Affecting the symbolic universe that the people you

want to communicate with live in; indirect effects; higher credibility for the communicator

• Difficult to clip, store and pass on, unlike online and press items (unless Youtubed)– Busy people tend to miss electronic messages

Communication• To key stakeholders or to media: it’s story telling

• Better have a single point, or at most 2-3 related points in a story; multiple stories don’t work at the same time

• Different stories appeal to different audiences/media; figure out what fits

• News releases give good training in working up narratives, but news releases/conferences less and less effective – Targeted dissemination; may need PR consultant in big

markets

Timing is everything

• Cannot release stories in the middle of elections/terrorist attacks, unless they have a connection

• Even outside CS schedule, if the opportunity arises, get your story out

• Map needs to be improved as you move through the jungle– Have the map, but don’t be its prisoner

Most important thing

• Audience• Identify your audience• Empathize with your audience• Address your audience’s needs• Test everything against the audience

If you take one thing away from this . .

• Audience is king– Identify your audience/interlocutor– Empathize with your audience/interlocutor– Respect your audience/interlocutor (especially

their time)– Address your audience’s/interlocutor’s needs– Test everything against the audience

Broadband Quality of Service Experience (QoSE)

Why Broadband QoSE?• The need for connectivity and growth in usage

- ~ 251 Million Internet subscribers in Asia Pacific of this ~128 Million Broadband subscribers– ITU 2007

• Lack of Quality - Complaints by users

- Broadband Quality of Service Standards not defined by many regulators

• Importance of Quality - Applications and platforms may require higher quality in connectivity for better performance

Methods of measurement

Existing methods - speednet • Emphasis on limited metrics – throughput

New Method• Methodology and software (AT-Tester) developed with IIT

Madras• Other than throughput it takes into account additional

metrics; Round Trip Time (RTT), Jitter and Packet Loss

Metrics explained

- Round Trip Time : Time taken for a packet to reach a destination and return

- Jitter : Average Variation of RTT

- Packet Loss :Number of packets (in %) which do not reach the destination

• Measures performance in 3 domains: ISP, national; International (yahoo.com)

InternetInternet

User

1st Entry point to US

Yahoo.com

ISP

National Server

3 Domains…

• Only Singapore regulator (IDA) states operator is responsible until the 1st point of entry to the US

Relevant metrics

+++ Highly Relevant ++ very relevant + relevant - not relevant

Throughput Delay

Service Down Up RTT Jitter Loss

Browse (text) ++ - + - -Browse (media) +++ - + + +Download file +++ - - - -Upload File - +++ - - -Transactions + + ++ + +Streaming media +++ - + ++ ++VOIP + + +++ +++ +++Games ++ + +++ ++ ++

Testing BBQoSE

• Testing conducted using AT-Tester Software- available at www.broadbandasia.net

• Testing in multiple domains- ISP, National and International (yahoo.com)

• Multiple broadband packages or links– 256kbps – 2Mbps

Testing contd…

• Multiple times a day– 6 times a day; ranging from peak to off peak

• Multiple Locations– Sri Lanka – Colombo– India – Delhi, Chennai (metro and outside)– Bangladesh – Dhaka

Results

Download from ISP & yahoo.com

Download from cloud (yahoo.com)

0%

40%

80%

120%

160%

200%

800 1100 1500 1800 2000 2300

TimeAc

tual

Spee

d as

a %

of st

ated

Sirius (256k) Dhaka, BD Airtel (256k) Delhi, INBSNL (256k) Chennai, IN Dialog (2M) Colombo, LKSLT (2M) Colombo, LK

Download from ISP

0%

40%

80%

120%

160%

200%

800 1100 1500 1800 2000 2300

Time

Actu

al Sp

eed a

s a %

of st

ated

Sirius (256k) Dhaka, BD Airtel (256k) Delhi, INBSNL (256k) Chennai, IN Dialog (2M) Colombo, LKSLT (2M) Colombo, LK

Return Trip Time to yahoo.com Yet to meet Singapore (IDA) standard

Time

ms

40

300 ms300 ms

Jitter – pinged to yahoo.com Almost within acceptable levels

41

Time

50 ms 50 ms 0

50

100

150

200

250

300

350

400

800 1100 1500 1800 2000 2300

Time

ms

Airtel (256k) Chennai, IN BSNL (256k) Chennai, IN

Dialog (2M) Colombo, LK SLT (2M) Colombo, LK

Importance of location…

Group work 1- 40 minutesSeparate into 3. Each is a policy research organization in

country X

You are attempting to influence 3 different stakeholders;• Group 1: Government• Group 2: Private Sector• Group 3: Non-governmental Organisations

• 7/8 members each• Appoint a moderator and speaker• Identify your audiences• Formulate a communications strategy for the

Broadband QoSE project

Government group

• Focus on policy and regulatory issues • May seek to convince policymakers and regulators give

higher priority to QoS issues. • It may not limit its activities to the AT tester, but may

want to implement additional quality testing methods.

Private-sector group

• Focuses on market mechanisms• May seek to influence Broadband Operators• Limited engagement with policy makers and regulators

NGO group

• Considers BBQoSE as key to good performance of all ICT-based NGOs

• May involve other NGOs in improving BB QoSE and implementing AT tester in multiple facilities

• May consider NGOs as proxies for the many disempowered citizens who use the Internet through their telecenters

• Involved in the testing process• May exert pressure on both Operators and on government

authorities